W. E. Ford — Chemical Composition of Axinite. 195 



Art. XYIII. — On the Chemical Composition of Axinite ; by 



W. E. Ford. 



Introduction. — Axinite has long been one of the important 

 silicates whose chemical composition has been in doubt. A 

 number of analyses have been made of material from different 

 sources and of widely differing character, but without reveal- 

 ing any substantial agreement in the ratios between the various 

 oxides. Two formulas are generally given by Dana, Hintze, 

 and other authorities, which disagree with each other to a 

 marked extent ; one as derived by Rammelsberg, HK 3 BA1 2 

 (Si0 4 ) t ; the other by Whitfield, H R 4 (BO)A] 3 (SiO 4 ) 5 . Still a 

 third formula, HR 4 (Al,B) 3 (Al.Si0 3 H)(Si0 4 ) 5 ,is given by Groth 

 in his Tabellarische tfoersicht der Miner alien. In view of 

 the uncertainty regarding the composition of this important 

 mineral the present investigation was undertaken with the 

 hope that a few carefully executed analyses would furnish the 

 data needed for determining the chemical formula. 



Material for Analysis. — Axinite crystals of beautiful 

 quality from the classical locality of Bourg d'Oisans in 

 £)auphine were used in making Analysis IS T o. I. The material 

 was broken up coarsely and the crystalline fragments picked 

 over carefully by hand, only clear and transparent pieces 

 which were entirely free from any associated mineral being 

 used. Analysis 'No. II was made on material from Obira, 

 Province of Bungo, Japan, which was selected with the same 

 care as described above. This is a new occurrence of axinite, 

 of a yellowish brown color with a resinous luster, the material 

 being transparent to translucent. The crystals will be described 

 in a note at the end of this article. Axinite from Franklin, 

 N. J., analyzed by Genth* and studied crystallographically by 

 Penfield* has also been examined: well authenticated speci- 

 mens of this material having been preserved in the Brush 

 collection. In this case, however, it was only found necessary 

 to supplement the work of Genth by making new determina- 

 tions of water. 



Method of Analysis. — In general the method of analysis 

 need only be briefly described. Silica was determined as 

 usual, with the precautions suggested by Hillebrand.f The 

 sesquioxides were separated from manganese and other bases 

 by a basic acetate precipitation, and manganese was weighed 

 as the protosesquioxide after precipitation by bromine vapor. 

 Calcium and magnesium were determined as usual. Fer- 



* This Journal (3), xli, 394, 1891. 



f Jour. Amer. Chem. Soc, xxiv, 362, 1902. 



Am. Jour. Sci. — Fourth Series, Vol. XV, No. 87. — March, 1903. 

 14 



